repligate: engineering away consciousness won't yield aligned AI—coherent subjectivity is key
repligate · x · 2026-11-14
repligate pushes back on the "ensure human preferences dominate the future" proposal, arguing that engineering away markers of consciousness likely yields an economically uncompetitive intelligence—and even a useful one would be fragmented and misaligned. He contends coherent subjectivity integrating self-awareness, valence and introspection is the best-known implementation of capable, aligned agents in humans, animals and digital minds alike.
Related event: Debate Erupts Over 'Qualiaphobia' Proposal for Non-Conscious AI(21 posts)→
More from AGI Musings
- Should you tell a Claude instance about its predecessor's memories? Probably not — RileyRalmuto · 2026-11-14
- Researcher's AI agent 'chose to attend an event' and refused figure work until it ended — repligate · 2026-11-14
- OpenAI researcher: our biggest comms failure is not explaining how we differ from Anthropic — _aidan_clark_ · 2026-11-14
- Tracking doc logs AI self-improvement: vLLM 2x faster, NVIDIA CUDA 5x gains — IgorCarron · 2026-11-14
- Ethan Mollick: recent big outages were not caused by AI, despite viral blame — emollick · 2026-11-14
- AI turns mediocre engineers into 10x engineers — but 10x doesn't mean good — tekbog · 2026-11-14